This project will develop multi-objective memetic algorithms and novel combinatorial
optimization techniques for core problems in bioinformatics. Major motivations include
genome-wide association studies and next-generation -omics technologies. An emphasis
will be on developing methods that address foundational issues related to practical
algorithmic utility. Implementations will include new hybrid and metaheuristic tools
that employ, whenever possible, mathematical programming and other efficient exact
solvers. The scalability of these algorithms will be studied in hybrid CPU-GPU systems,
and quantified by their empirical performance on extremely large datasets. A key aim
is to gain at least two orders of magnitude in run-time improvement over what can be
addressed with current systems and algorithms.
